Method and apparatus for measuring the quality of an image

Abstract

An electronic device that includes an image quality measuring function and an operation method thereof are disclosed. An image quality measuring method includes: obtaining an image; classifying an image scene category of the image; determining a classifier corresponding to the classified image scene category; determining image quality factor scores with respect to the image; and evaluating image quality with respect to the image by using the determined image quality factor scores and the determined classifier.

What is claimed is: 
     
         1 . An electronic device, comprising:
 a memory that stores a plurality of images and a plurality of classifiers; and   a processor electrically connected to the memory,   wherein the processor is configured to at least:   analyze a category of an image of which an image quality evaluation is requested, and determine a classifier corresponding to the category of the image from among the plurality of classifiers;   determine image quality factor scores of the image; and   evaluate the image quality of the image based on the determined image quality factor scores and the determined classifier.   
     
     
         2 . The electronic device of  claim 1 , wherein the processor comprises:
 an image factor extracting module configured to measure a quality for each image quality factor of the image;   a category classifying module configured to classify an image scene category of the image;   a classifier selecting module configured to select an image quality classifier corresponding to the image scene category; and   an image quality evaluating module configured to determine a total image quality score of the image based on results from the image factor extracting module and the classifier selecting module.   
     
     
         3 . The electronic device of  claim 2 , wherein the category classifying module is configured to analyze the image scene category based on a scheme that classifies an image using feature vectors of the image or a scheme that classifies an image using learning. 
     
     
         4 . The electronic device of  claim 2 , wherein the classifier selecting module is configured to select a classifier corresponding to the image scene category from among image quality classifiers stored in advance in the memory. 
     
     
         5 . The electronic device of  claim 2 , wherein the image factor extracting module is configured to divide the image into a plurality of areas, and to measure a blur factor score for each divided area. 
     
     
         6 . The electronic device of  claim 5 , wherein the image factor extracting module is configured to:
 measure a blur factor score for each of the divided areas of the image, and determine, as a target area of measurement, one or more areas that have a high sharpness out of the divided areas,   wherein the target area of measurement includes a predetermined area or a group of a predetermined number of adjacent areas of the divided areas, having a low blur factor score.   
     
     
         7 . The electronic device of  claim 2 , wherein the image factor extracting module is configured to extract, from the image, a distribution of edge components for each direction, and to measure a blur factor score for image quality evaluation based on an area that has a large difference in cumulative edge distributions from among areas where a horizontal edge component and a vertical edge component are perpendicular to each other. 
     
     
         8 . The electronic device of  claim 2 , wherein the image quality evaluating module is configured to extract a likelihood score corresponding to the image, using image quality factor scores that are transferred from the image factor extracting module and the image quality classifier transferred from the classifier selecting module, and to determine a total image quality score with respect to the image based on the extracted likelihood score. 
     
     
         9 . The electronic device of  claim 2 , wherein the processor comprises:
 an image managing module configured to obtain the image from the inside or the outside, and to recognize image information for measuring the quality of the image.   
     
     
         10 . The electronic device of  claim 2 , wherein the processor comprises:
 a category determining module including processing circuitry, the category determining module configured to determine the image scene category with respect to the image, using the image scene category transferred from the category classifying module and the total image quality score transferred from the image quality evaluating module.   
     
     
         11 . A system for measuring a quality of an image, the system comprising:
 a first electronic device configured to request image quality evaluation of an image; and   a second electronic device operatively connected to the first electronic device,   wherein the second electronic device is configured to at least:   obtain the image from the first electronic device;   analyze a category of the image, and determine a classifier corresponding to the category of the image;   determine image quality factor scores of the image; and   evaluate an image quality of the image based on the determined image quality factor scores and the determined classifier, and provide the first electronic device with a result of the image quality evaluation.   
     
     
         12 . A method of measuring a quality of an image, comprising:
 obtaining an image;   classifying an image scene category of the image;   determining a classifier corresponding to the classified image scene category;   determining image quality factor scores with respect to the image; and   evaluating an image quality with respect to the image, using the determined image quality factor scores and the determined classifier.   
     
     
         13 . The method of  claim 12 , wherein the classifying comprises:
 analyzing the image scene category based on a scheme that classifies an image using feature vectors of the image and a scheme that classifies an image using learning.   
     
     
         14 . The method of  claim 12 , wherein the determining of the classifier comprises:
 selecting a classifier corresponding to the image scene category from among various image quality classifiers stored in advance.   
     
     
         15 . The method of  claim 12 , wherein the determining of the image quality factor scores comprises:
 dividing the image into a plurality of areas;   measuring a blur factor score for each of the plurality of divided areas;   determining an area that has a low blur factor score; and   measuring an image quality based on the determined area.   
     
     
         16 . The method of  claim 15 , wherein the determining of an area having a low blur factor score comprises:
 determining, as a target area of measurement, one or more areas of the divided areas of the image having a high sharpness,   wherein the target area of measurement includes a predetermined area or a group of a predetermined number of adjacent areas of the divided areas, which has a low blur factor score.   
     
     
         17 . The method of  claim 12 , wherein the determining of the image quality factor scores comprises:
 extracting, from the image, a distribution of edge components for each direction;   determining a cumulative edge distribution based on the extracted distribution of edge components for each direction;   determining a target area of decision that includes the largest number of edge components;   determining, from the target area of decision, a target area of measurement based on a difference in cumulative edge distributions; and   measuring an image quality based on the target area of measurement.   
     
     
         18 . The method of  claim 17 , wherein the determining of the target area of measurement comprises:
 determining the target area of measurement based on a difference in cumulative edge distributions from among areas where a horizontal edge component and a vertical edge component are perpendicular in the target area of decision.   
     
     
         19 . The method of  claim 12 , wherein the determining of the image quality score comprises:
 classifying an image as a high quality image class or a low quality image class using the image quality factor scores and the classifier;   extracting a likelihood score corresponding to the image; and   determining a total image quality score with respect to the image, based on the extracted likelihood score.   
     
     
         20 . A computer readable recording medium that stores a program which, when executed in an electronic device, causes the electronic device to perform operations comprising:
 classifying an image scene category of an image and determining a classifier corresponding to the classified image scene category;   determining image quality factor scores with respect to the image; and   evaluating an image quality with respect to the image, using the determined image quality factor scores and the determined classifier.